TVB

Pa

b

c

d

a

ARAA

KPKP

1

tkttnossntdtt

W

fg

1h

NJAS - Wageningen Journal of Life Sciences 67 (2013) 1– 10

Contents lists available at ScienceDirect

NJAS - Wageningen Journal of Life Sciences

jou rn al h om epage: www.elsev ier .com/ locate /n jas

he Impact of the Product Generation Life Cycle on Knowledgealorization at the Public Private Research Partnership, the Centre forioSystems Genomics

.J.P. Garbadea, S.W.F. Omtab,∗, F.T.J.M. Fortuina, R. Hall c,d, G. Leonec

Food Valley Organization, P.O. Box 294, 6700 AG Wageningen, The NetherlandsWageningen University, P.O. Box 8130, 6700 EW Wageningen, The NetherlandsCentre for BioSystems Genomics, P.O. Box 98, 6700 AB Wageningen, The NetherlandsPlant Research International, B.U. Bioscience, P.O. Box 16, 6700AA, Wageningen, The Netherlands

r t i c l e i n f o

rticle history:eceived 4 May 2012ccepted 5 July 2013vailable online 25 October 2013

eywords:ublic private research partnershipsnowledge valorization

a b s t r a c t

The present paper aims to address the impact of the product generation life cycle (PGLC) on knowledgevalorization in public private research partnerships (PPRPs). Data were collected from participants in theCentre for BioSystems Genomics (CBSG), a Dutch PPRP program in the plant breeding sector. In total, 15commercial partners participated in the study, 7 with a relatively short PGLC of 5 to 6 years, active in thetomato sector, and 8 potato partners, having a very long PGLC of up to 25 years. The results show a clearrelationship between CBSG’s valorization support activities and the level of knowledge utilization by theparticipants, although the preferred type of support activities differs between the potato and tomato

roduct generation life cycle companies. Firms with a long PGLC, having a higher complicacy of the R&D process, require more basicresearch support and extra communication tools that help to bridge the gaps caused by the long durationof the development process. Companies with short PGLCs, being challenged to keep development timeof new products as short as possible in order not to miss out on market opportunities, value the PPRPmost for the networking possibilities and as provider of the latest technological developments.

Roya

© 2013

. Introduction

Stimulating innovation stands high on national and suprana-ional political agendas. Innovation involves the conversion of newnowledge into a new product, process or service and bringinghis new product, process or service into use [1]. Since innova-ions are increasingly being established within inter-organizationaletworks [2,3] and resulting from recent success stories of so-calledpen innovation [4], governments are searching for new ways totimulate innovation by involving the public and private sector andtimulating partnerships between them [e.g. 5]. Public private part-erships (PPPs) aim to combine “the resources of government withhose of private agents (business or not-for-profit bodies) in order to

eliver societal goals” [6]. Since the resources of government includehe publicly-financed research organizations, knowledge is one ofhe main resources that is brought into the partnerships from the

∗ Corresponding author. Wageningen University; P.O. Box 8130, 6700 EWageningen, The Netherlands; Tel.: +31318655965.

E-mail addresses: onno.omta@wur.nl (S.W.F. Omta),rances.fortuin@foodvalley.nl (F.T.J.M. Fortuin), robert.hall@cbsg.nl (R. Hall),ionata.leone@cbsg.nl (G. Leone).

573-5214/$ – see front matter © 2013 Royal Netherlands Society for Agricultural Sciencttp://dx.doi.org/10.1016/j.njas.2013.07.002

l Netherlands Society for Agricultural Sciences. Published by Elsevier B.V. All rights reserved.

public side to transform it into value for society [7]. Perkmann andWalsh [8] introduce respectively the term public private researchpartnership (PPRP) that will be used throughout this paper.

PPRPs aim, to enhance knowledge valorization by the participat-ing public research organizations and utilization by participatingcompanies, which is by definition, the formal transfer of knowl-edge resulting from basic or applied research to the commercialsector for economic benefit [9]. In the literature examples of largePPRPs are found, such as SEMATECH, established in the USA with100 Million dollars of federal funding in 1987 to regain a lead-ing position in computer manufacturing by combining private andgovernmental know-how [10]. Recent studies show a widespreaduse of university-industry partnerships in Austria [11], the UnitedKingdom [12] and Germany [13]. However, there are also some con-cerns about the effectiveness of PPRPs. Adams [14] found a timelag of approximately 20 years between starting research and themoment that industry can profit. Geisler [10] argues that gains fromPPRPs appear mainly to lie in leveraged R&D rather than in the num-ber of product innovations, while Feller [15] claims that firms, by

establishing relationships with universities, aim for generic bene-fits, such as coming into contact with young researchers who arepossible future employees, rather than to commercialize scientificinnovations. Although these generic benefits are important, these

es. Published by Elsevier B.V. All rights reserved.

dx.doi.org/10.1016/j.njas.2013.07.002www.elsevier.com/locate/njashttp://crossmark.crossref.org/dialog/?doi=10.1016/j.njas.2013.07.002&domain=pdfmailto:onno.omta@wur.nlmailto:frances.fortuin@foodvalley.nlmailto:robert.hall@cbsg.nlmailto:gionata.leone@cbsg.nldx.doi.org/10.1016/j.njas.2013.07.002

2 en Jou

ctqcgaitpuscfiPria

Cga

oddctma

2

2

btpfwsopetpiafie[trmpbrbTnruu

uct generation life cycle (PGLC) as being the sum of the productlife cycles of all related products belonging to one product genera-tion. Across industries, huge differences in the average length of the

P.J.P. Garbade et al. / NJAS - Wagening

oncerns bring up the research question: ‘Do PPRPs really increasehe level of knowledge utilization by companies?’ An additionaluestion is, whether the effectiveness of knowledge utilization byompanies in PPRPs can be expected to be independent of contin-encies or instead, are dependent upon certain parameters, suchs company size [16,17] or type of industry sector [18]. A furthernteresting parameter to be taken into account is the length ofhe product generation life cycle (PGLC), which is the sum of theroduct life cycle of all related products belonging to one prod-ct generation. Fortuin [19] identified the PGLC in a cross-industrytudy as the dominant factor affecting the entire innovation pro-ess, from knowledge generation up to market introduction of thenal product. This raises the question whether the length of theGLC also has an impact on the effectiveness of PPRP’s. One relatedesearch question consequently is: ‘Does the PGLC length of partic-pating companies influence the knowledge valorization process in

PPRP?’The present study investigated these research questions in the

entre for BioSystems Genomics (CBSG), a Dutch PPRP in plantenomics, involving breeding companies active in typically longs well as short life cycle products.

The paper is structured as follows. Section 2 describes the the-retical foundation of the study which provides the basis for theevelopment of a conceptual model. Section 3 describes the studyomain, section 4 the development of a survey, the methods of dataollection, analysis used and the operationalization of the concep-ual model. Section 5 presents the results of the survey of the CBSG

ember companies. Finally, in section 6, the results are discussednd conclusions are drawn.

. Theoretical Framework

.1. Knowledge valorization

Knowledge is generatedin both public and private organizationsut is driven by different motivations in the different organiza-ions. For private organizations,the economic needs and/or therofit orientation can be assumed to play a major role. So theyocus more on applied research and the exploitation of knowledge,hich is by definition concerned with the refinement and exten-

ion of existing technologies [20]. Public research in contrast, is freef economic needs, although this view can be questioned[21]sinceublic research institutions are increasingly being judged on theirconomic performance.Up to 25% of academic research is expectedo be influenced directly by industrial funding [22]. The fact thatublic research is largely financed by national or supranational

nstitutions such as the EU, gives the research a certain directionnd tries to align it with the needs of society. In most cases it has aundamental character and therefore, is related to exploration andnvestigation, which is rooted in the quest for potential new knowl-dge [23]that can help tackle previously unresolved problems24].Entrepreneurship is the most important factor concerning theransfer ofthis new knowledge to the market [25].Exploitation ofesults by developing new products and bringing them to com-ercial markets istherefore, expected to be conducted better in

rivate organizations.Knowledge valorizationoffers the tools forridging the gap between exploration and exploitation of researchesultsand therefore, it seems there are evident advantages in com-iningthese strengths of public and private organizations inPPRPs.his applies especially to science-based sectors such as biotech-

ology.These show high complementarities between academicesearch and commercial R&D [7], a high importance status ofniversity generated IP [26] and provision of company staff byniversities[21,27].

rnal of Life Sciences 67 (2013) 1– 10

Access to networks and proximity of knowledge are key to trig-ger innovation (Piet Schalkwijk, IPR director of Akzo Nobel, 20101).Therefore, to be excluded from knowledge access couldmean alarge competitive disadvantage for companies. In a competitivebusiness environment a potential risk for innovationoccurs whensome (especially small) companies are excluded from the networkand therefore from knowledge access. PPRPs can increasea com-panies’ innovative reputation [28]and balance out a knowledgeaccess blockage that companies might face by allowingalso smallercompaniesto participate andthus benefit from the knowledgecreated.Consequently, in science-based sectors, manyfirmsstimu-lateindustrial researchers to interact with academia, and also joinforces to generate sufficient critical mass[29].

One form of knowledge valorization in PPRPs includes agree-ing upfront about the companies’ right to use the research resultsfor commercial exploitation for a certain license fee [15]. Typi-cally, these fees are limited as both the public (universities andresearch institutes) and the private (company) partners contributeto the research. While knowledge is normally published by pub-licly financed researchers without claiming any exploitation rightsto it [30], in PPRPs knowledge exploitation rights may need to bedistributed according to the partners’ contribution to the researchor relating to monetary issues. This needs to be justified to thetaxpayer, who does not have an interest to finance research andtransfer the rights of research results at a low cost price to one orseveral exploiting parties, aiming to generate profits from them.This justification can be found in the gains in the exploitation effi-ciency of the generated knowledge, so that society in general, canfinally benefit from a technology or product that would not bedeveloped otherwise.

This efficiency increase or net economic benefit has however,yet to be proven [25], so in order to justify the potential privatiza-tion of formerly public knowledge, this step has at least to result ina better performance of the companies participating in PPRPs, com-pared to companies that do not. If intellectual property rights (IPRs)are supposed to show value in the knowledge valorization process,they are supposed to generate, besides the exclusive exploitationrights to the organization owning them, also a broader knowledgebase to conduct further research for the whole research commu-nity. This can be achieved when the IPRs are vested at the publicresearch institutes. In the specific case of plant breeding, wherethe IPRs are granted in the form of plant breeders rights (PBRs), theresearch exemption would also grant such an additional benefit tothe researchers’ community, since the PBR holder is always obligedto provide samples of the protected plant varieties to parties plan-ning to conduct further research on them.

2.2. Impact of the length of the product generation life cycle(PGLC)

The product generation life cycle (PGLC) is built on the well-known concept of the product life cycle (PLC). Bayus [31] definesthe product life cycle as the evolution of unit sales over the entirelifetime of a product. The product life cycle [e.g. 32, 33-35] has fourstages: introduction (an initial period of slow sales growth), growth(a period of rapid growth in sales), maturity (a period in which saleslevel off and are relatively stable), and decline (when sales dropoff). Maidique and Zirger [36] introduced the concept of the prod-

product generation life cycle (PGLC) of products can be observed,

1 Own observation 2010.

P.J.P. Garbade et al. / NJAS - Wageningen Jou

Kno wled ge utilizati on

performance

Produc t generation li fe

cycle*

Access

Support acti vities

Network growth

Knowledge valorization support

Fs*

rcar

cpopmilosdt

aPoskitsFucc

lcit

Fppvta

3

3

rcgaa

igure 1. Conceptual model: Knowledge valorization in a public private partner-hip.PGLC proxy: tomato has a short, potato has a long PGLC.

anging from less than one year, such as in the mobile phone andomputer industry, to over 20 years, such as in the pharmaceuticalnd aircraft industries [37]. Fine [38] and Brown and Eisenhart [39]efer to these differences as ‘industry clock speed’.

In a cross-industry study of 10 multinational technology-basedompanies Fortuin [19] investigated the impact of the length of theroduct generation life cycle on the innovation process. The lengthf the PGLC proved to have a major impact on the entire innovationrocess from the knowledge generation in basic research up to thearket introduction of the final product. In typically long life cycle

ndustries, companies are generally confronted with high techno-ogical complexity, leading to an elongation of the research partf the innovation process. Companies in industries with relativelyhort PGLCs are typically confronted with a high level of marketynamism and competition leading to extra pressure to speed uphe R&D process in order to shorten the time-to-market.

Companies with shorter PGLCs are more bothered by the speedt which things develop compared to companies with longerGLCs. Receiving information in time and not to miss out onpportunities should play a much more prominent role, which isupposed to determine a high importance of knowledge access andnowledge transfer within a PPRP structure. Since public researchnstitutes build on the curiosity and economical independency ofheir researchers, they represent a valuable pool to tap for a diver-ity of up-to-date information on a wide range of knowledge fields.or companies with a long PGLC, the possibility to reduce the prod-ct development time by involving publicly generated knowledgeould be an effective way to valorize knowledge generated in a PPRPontext.

In the plant breeding industry, major differences exist in PGLCength. These will be discussed for potato and tomato breedingompanies in Section 3. Since both types of company participaten CBSG, this provides a unique opportunity to study the impact ofhe length of the PGLC on valorization within one industry sector.

Our hypothesis leads to the conceptual model as represented inigure 1 below. In this model the knowledge valorization support asrovided by the PPRP is conceptualized as consisting of access, sup-ort activities and network growth enhancement. The knowledgealorization support is supposed to result in a knowledge utiliza-ion performance. In this relationship, the PGLC is supposed to have

mediating effect.

. Study Domain

.1. The Centre for BioSystems Genomics (CBSG)

In the Netherlands the traditional agricultural knowledge valo-ization model, has been based on co-financed research initiatives

onnecting public and private research. From the 1990s, the Dutchovernment initiated the set-up of networks of public researchnd industry organizations in specific technology areas [40]. As

consequence the old model has been increasingly replaced by

rnal of Life Sciences 67 (2013) 1– 10 3

establishing PPRPs on research programs [41]. CBSG is an exam-ple of a Dutch PPRP in plant genomics, including 4 universities andtwo research institutes, 6 vegetable seed breeding companies, 5potato breeding, 1 potato processing company, 1 genomics tech-nology company and 2 potato commodity boards. CBSG aims toexploit the full potential of a broad range of genomics approachesin order to create new opportunities for sustainable crop produc-tion, enhanced food quality with reduced environmental impact.Research is focussed around a fully-integrated research programmetargeting potato, tomato and, to a less extent, Brassica crops. CBSGwas established in 2002 with a total research budget of 53 MDfor 5 years. In 2008, CBSG entered its second 5 year period withan equivalent budget. Some 15% of the CBSG total budget is paidby the private partners. CBSG carries out plant genomics researchusing the latest, state-of-the-art technologies. The limited choice ofcrops has deliberately been made to maintain focus and to cover thespecies of greatest importance for the Dutch agro-food industry.

3.2. Public private research partnerships (PPRPs) in the breedingindustry

Collaborative research and informal contacts are more impor-tant in the plant breeding field compared to other, more appliedfields of technology production [13]. Therefore it is not surprisingto find Dutch plant breeding companies engaging in a number ofdifferent PPRPs with knowledge institutions. A PPRP, such as CBSG,has to serve different stakeholders. The knowledge institutions, thecompanies and, last but not least, society have to benefit from thePPRP. The knowledge institutions expect that excellent science willtake place, resulting in a large number of highly rated scientificpublications, which is the primary performance indicator they aremeasured against. They also expect the PPRP to result in new con-tract research, bilateral research agreements and last but not least,also in the generation of extra income by licensing out their intel-lectual property rights (IPRs) to industry. Industry in turn expectsthat their participation in a PPRP results in tangible products in theform of tools, methods, and products etc which they can use, asschematically presented in Figure 2. Society (tax payers and gov-ernment) expects results both in science and education (includingtraining high quality researchers and PhDs), as well as valorizationin terms of new, improved products that are important for societyand induce extra employment.

3.3. Tomato versus potato breeding

Many of the world’s leading plant breeding companieshave their headquarters and/or important R&D facilities in theNetherlands [42]. The CBSG partners are companies that are inthe top 10 tomato seed companies operating worldwide. The mainDutch potato companies operate in the Dutch, European, and globalpotato markets. While some tomato seed companies are also pow-erful multinationals, they all develop seeds not only for tomato, butalso for other important vegetables, e.g. for cucumber, cauliflowerand pepper. The potato breeding companies focus instead only onseed potato production. The partners in CBSG cover ca. 85% of theglobal fresh tomato seed market and 75% of seed potato production.

The differences between potato and tomato breeding compa-nies start from the biological differences in the plants, especially inthe way the crops are propagated and cultivated. Diploid tomatocultivars require 3 to 8 years to be bred which allows them toreach their commercial peak before 25 years (plant breeding pro-tection time). Tomato seeds are sold as F1 hybrids, implicating that

the next generation will not inherit the same traits, so F1 hybridscan work as IP protection. Due to the complex tetraploid genome,potatoes require 10 to 20 years to be bred and propagated readyfor release. Twenty five years is therefore too short for a potato

4 P.J.P. Garbade et al. / NJAS - Wageningen Journal of Life Sciences 67 (2013) 1– 10

tems G

cetcaiak

dpFdiast

4

amcr(cFimtRuaqpap[

vozhW

Figure 2. Centre for BioSys

ultivar to be profitable, so plant breeders rights (PBRs) have beenxtended to 30 years. For potatoes, molecular breeding possibili-ies and therefore the steering capacity of the breeding process isurrently limited compared to tomato molecular breeding. Tomatond potato companies show consequently a tremendous differencen PGLC length, which makes them an ideal study population tonswer our research question ‘Does the PGLC length influence thenowledge valorization process in a PPRP?’

In the Netherlands, tomatoes are grown under controlled con-itions in greenhouses. Therefore tomatoes are more readilyrotectable and hence, are subjected to lower disease pressures.urthermore, many current varieties already carry importantisease resistances and therefore, breeding can focus more on qual-

tative traits: taste, fragrance and appearance. Potatoes in contrast,re vegetatively-propagated and cultivated in open fields. As a con-equence, potato breeding is strongly directed towards resistanceo devastating diseases such as late blight (Phytophthora).

. Methods of Data Collection and Analysis

In 2009 a research project was conducted with the aim ofssessing how effective CBSG has been in valorizing its funda-ental and applied research results in which all CBSG member

ompanies participated. As done by Perkmann and Walsh [8], theesearch is not based on publically available intellectual propertyIP) databases but on a company survey used so as not to miss out onollaborative aspects and results that do not occur in IP databases.rom the companies, interviewees were selected based on theirnvolvement in CBSG: they were contact person, project leader, or

ember of the CBSG management team. Within their organizationshe participants fulfill one or more roles as: researcher, breeder,&D manager, or director. The valorization of knowledge was eval-ated by means of a 207 item questionnaire. The questions were

mix of closed questions that used 7-point Likert scale, and openuestions in which quantification of the CBSG’s valorization sup-ort was requested. There were questions regarding CBSG accessnd support, frequency of use of CBSG services and CBSG relatederformance to link PPRP specific inputs and outcomes directly43]. The questions used are given in Table 1.

In total, 15 questionnaires were analyzed, one for each pri-ate partner. These have been categorized according to the typef industry, place in the value chain, and the size of the organi-ation. Since the data are non-parametric, the questionnaire inputas been analyzed by using Spearman rank correlation and Kruskalallis–tests.

enomics’ main objectives.

5. Results

5.1. Baseline description and Centre for BioSystems Genomicsoutput

Seven of the participating companies belong to the tomatoindustry and eight belong to the potato industry. Thirteen com-panies (7 tomato companies and 6 potato companies) are directlyinvolved in breeding, while one potato company has a daughtercompany conducting the breeding activities. Two partners partic-ipating in CBSG have their core activities in processing. Twelveorganizations (7 tomato companies and 5 potato companies) arelarge firms (annual sales > 100MD ), and three (potato companies)are small firms (annual sales < 50MD ). The list of participating orga-nizations can be found on the CBSG public website: www.cbsg.nl

Figure 3 shows some output of CBSG in the period 2002-2012.CBSG was considerably successful in terms of its scientific output,with more than 750 scientific publications, including papers in Sci-ence and Nature and more than 70 successful PhD defenses. Interms of knowledge valorization, 16 patents were filed, 23 licenseswere awarded and three spin off company was established. In theinterviews, six companies indicated that up to four new prod-ucts could be developed as a result of the CBSG activities. Up to100 tomato/potato genetic markers potentially useful for breedingcould be tested and are expected to be implemented. One companyfurther indicated that CBSG participation could lead to 10 new vari-eties. Another company indicated that 90% of future products willbe derived from CBSG activities.

5.2. Knowledge valorization at Centre for BioSystems Genomics

To shed light on the knowledge valorization and the effect of thePGLC length, the results as reported in Table 2 are discussed. First,the means of a number of answers given by tomato and potato com-panies with focus on significant differences between the two typesof companies are compared by applying a Kruskal Wallis test ontwo independent samples (vertical dimension in Table 2). Then, theparameter ‘knowledge valorization performance’ has been relatedto the parameter ‘knowledge valorization support’ activities. Bylooking at the correlations found for the potato and tomato com-panies separately, it has been extrapolated where both companytypes show a similar pattern and where there are differences

that can be related to the PGLC length (horizontal dimension inTable 2).

When looking at the mean differences of potato and tomatocompanies in terms of knowledge valorization performance,

P.J.P. Garbade et al. / NJAS - Wageningen Journal of Life Sciences 67 (2013) 1– 10 5

Table 1Summary of variables used in the questionnaire.

Conceptual Level (parameters) Indicators Please indicate the extent to which youagree with the following statements: Likertscales from 1 to 7, while 1 = completelydisagree, 7 = completely agree

Knowledge valorization performance Basic research By participating in the CBSG program myorganization expects to improve the basicresearch process.

Plant breeding By participating in the CBSG program myorganization expects to improve its breedingprocess.

Successful research completion Participating in the CBSG increases the chancesof successful research completion

Breeding strategy Participating in the CBSG program enables myorganization to improve its breeding strategy.

Breeding time reduction Please quantify the percentage of the breedingprocess time reduction (%)

Tested markers Participating in the CBSG program enables myorganization to increase the number oftomato/potato markers that will be tested.

New products developed Participating in the CBSG program enables myorganization to develop new products.

New products launched By participating in the CBSG program myorganization expects to launch new productsto the market.

Valorization level In general which grade would you assign to thevalorization of the CBSG research resultingfrom the participation in the CBSG program? 1:very low; 7: very high

Strengthened image Participating in the CBSG program enables myorganization to strengthen its image

Knowledge and skills Participating in the CBSG program enables myorganization to improve the level ofknowledge and skills of the personnel.

Below per indicator two questions are asked regarding CBSG support (unless otherwise introduced).1.) How important is for your organization? 1 = not important; 7 = very important2.) Which value describes best the frequency of use 2003-2008? 1 = never, 2 = once per 3 years, 3 = once per year, 4 = once per quarter, 5 = once per month,6 = once per week, 7 = daily

Knowledge valorization support Access Contact with CBSG researchers Contact with qualified CBSG researchersIntranet Access to CBSG intranet informationDatabases Access to databasesAccess to IP Help with getting access to Intellectual

Property (licenses, plant breeders rights,patents)

International research programs Access to information on internationalresearch programs

Summit Access to the annual CBSG summitInfrastructure Access to infrastructure (e.g. R&D labs,

equipment, instruments)Support activities Technology monitoring Help with new technology monitoring and

road mappingTechnology advice Access to technology adviceIP filing Help with Intellectual Property filing (plant

breeders rights, patents)Bio-informatics Access to bio-informatics knowledge and

servicesTroubleshooting Help with troubleshootingSharing R&D costs Possibility to share R&D costs with other

companies (conducting research collectivelywith CBSG partners)

Recruiting new researchers Participating in the CBSG program enables myorganization to recruit new researchers orassistants (1: completely disagree; 7completely agree)

Network growth Company interaction Enhanced interaction with other companies inthe potato/tomato sector

Bilateral research Help to set up bilateral research program withother CBSG partners

Has CBSG support led to extra researchwithin the CBSG framework?Extra research with knowledgeinstitutions

Percentage extra research with knowledgeinstitutions within CBSG (%)

Extra research with commercialpartners

Percentage extra research with othercommercial partners within CBSG (%)

6 P.J.P. Garbade et al. / NJAS - Wageningen Journal of Life Sciences 67 (2013) 1– 10

enom

TeoppressplstcptTpPatboic

inltapcritWcpoiebsm

Figure 3. Centre for BioSystems G

able 2 (vertical dimension in table) shows a significant differ-nce between the results found for the indicators ‘improvementf the basic research’, ‘plant breeding’ and ‘breeding strategy’. Theotato breeders agree with the statement that from their partici-ation in the CBSG program, they expect an improvement in basicesearch and plant breeding, and state that CBSG participation hasnabled a breeding strategy improvement. The tomato companieslightly disagree with these statements. These results should beeen within the perspective of the learning potential towards theserocesses, since the tomato companies already have more molecu-

ar breeding facilities in house. The higher importance of breedingtrategy improvement for the potato companies seems also linkedo the higher complexity of their own breeding process. When itomes to the indicator ‘time reduction in the breeding process’,otato breeding companies indicate any reduction of 19%, whilehe tomato breeding companies did not indicate a time reduction.his could be explained by the fact that tomato breeding com-anies are already close to the minimum, biologically-possible,GLC length, with further time reductions being much harder tochieve. Another significant difference between the potato andomato breeding companies is the increase in the indicator ‘num-er of markers’ that are tested due to CBSG participation. Forther knowledge valorization performance indicators, no signif-cant mean differences were found between tomato and potatoompanies.

Concerning differences in the knowledge valorization supportn Table 2 (see also Table 1 for explanation), at first it should beoted that the parameter ‘access’ to CBSG was evaluated simi-

arly by potato and tomato companies. Potato companies judgehe importance of the indicator ‘intranet access’ quite high withn average score of 6.3 on a scale from 1 to 7, while tomato com-anies score this slightly lower with an average of 5.9. The tomatoompanies judge the importance of indicator ‘contact with CBSGesearchers’ to be as important as the indicator ‘access to the CBSGntranet’, while the potato companies give a slightly lower scoreo the indicator ‘importance of contact with the CBSG researchers’.

hen looking at the indicator ‘frequencies of the CBSG access’, theontact frequency between CBSG researchers and the tomato andotato companies is almost the same, between once per quarter andnce per month (data not shown). Potato companies use the CBSGntranet more than once per month, tomato companies about once

very two months (data not shown). The CBSG database is used byoth company types, once per quarter (data not shown). The onlyignificant difference concerns the indicator ‘annual CBSG sum-it’, where tomato companies judge it as important with a score of

ics 2002-2012 output overview.

5.7 while the potato companies give it a 3.8. When evaluating theparameter ‘support activities’ provided by CBSG (see also Table 1),the potato companies give a higher importance to indicator ‘tech-nology monitoring and road mapping’ and use it significantly moreoften than the tomato companies. At the same time, the potatocompanies also accept more frequently the help of CBSG when itcomes to the indicator ‘troubleshooting’ and judge the importanceof the indicator ‘R&D cost sharing’ higher than tomato companies.The importance of the indicator ‘R&D cost sharing’ seems to be inline with the findings of Fortuin [19] in that with long PGLC com-panies, the R&D costs are higher and have to be spread out overa longer time period due to facing higher levels of uncertainty.Tomato companies, with a shorter PGLC, in contrast, value morehighly the access to the annual CBSG summit and the CBSG website,which can be related to the fast changing short life cycle innova-tions. Here, being updated in time on the very latest developmentsis crucial.

Concerning the effect of the parameter ‘network growth’(Table 1), the potato companies show a higher frequency withregard to the indicator ‘interacting with other companies’ - aboutevery two months (data not shown), while the tomato companiesassess this item to occur about once per year (data not shown).Potato companies indicate also 2.4% extra research with other CBSGcommercial partners, while the tomato companies state no extraresearch with regard to this indicator.

To see whether the differences found between potato andtomato companies also impact on the role that certain factors havein our valorization model, an in depth analysis was conducted.Therefore, the Spearman rank correlations for tomato and potatocompanies have been examined separately (horizontal dimensionin Table 2), throughout all the indicator questions. The expecta-tions towards the indicator ‘basic research improvements’ correlatewith the indicator ‘successful research completion’, for both potatoand tomato companies. For the potato companies, the high corre-lation of 0.95** between the indicator ‘basic research’ and ‘plantbreeding’ suggests that they see the indicator ‘plant breeding’ asclosely related to ‘basic research’. For both the potato and tomatocompanies the expectations towards the indicator ‘plant breed-ing’ correlate with ‘breeding strategy improvement’, but correlatesonly for the potato companies with the indicator ‘expectationsto launch new products to the market’. The increase in chances

of ‘successful research completion’ correlates for both potato andtomato companies with indicator ‘improvement of the level ofknowledge and skills’ of their personnel as well as with the indi-cator ‘expectations to launch new products to the market’. For

P.J.P. G

arbade et

al. /

NJA

S -

Wageningen

Journal of

Life Sciences

67 (2013) 1– 107

Table 2Mean and standard deviation and significant Spearman rank correlations of potato (P, n = 8) and tomato (T) companies (n = 7)a

Indicators Mean (Stdv) Knowledge valorization performance

1 2 3 4 5 6 7 8 9 10 11

1 Basic research P 5.3* (1.8) xT 3.4 (1.7) x

2 Plant breeding P 5.3* (1.9) .95** xT 2.2 (1.5) x

3 Successful researchcompletion

P 3.9 (1.6) .56 x

T 4.1 (1.7) .64 x4 Breeding strategy P 5.3* (1.8) .69 .75* x

T 2.7 (1.9) .62 .80* x5 Breeding time reduction P 19%(18%) .89* .95** .80 x

T No time red. x6 Tested markers P 6.3* (0.5) .66 X

T 3.7 (2.4) X7 New products developed P 4.3 (2.1) .67 x

T 4.0 (2.2) .64 x8 New products launched P 4.5 (1.8) .73* .64 .67* .94** .85* .69* X

T 3.8 (2.3) .67 .62 .99** X9 Valorization level P 4.6 (1.5) .75* .78* .72 .67 .73* .85** x

T 3.6 (1.5) .59 .82* .62 x10 Strengthened image P 4.5 (1.6) .83** .72 .70* .69 .61 .76* .93** x

T 4.0 (2.2) .77* x11 Knowledge and skills P 6.0 (0.7) .78* .60 x

T 4.7 (1.9) .66 x

Indicator Mean (Stdv) Knowledge valorization support

1 2 3 4 5 6 7 8 9 10 11

Contact with CBSG researchers (importance) P 5.0 (1.7) .84* .87* .86* .68*

T 5.8 (0.9) .64 .56 .73* .75* .86* .85*

Contact with CBSG researchers (frequency) P 4.4 (0.9)T 4.3 (0.5) -.89* -.87*

Intranet (importance) P 6.3 (0.7) .61T 5.9 (1.1)

Intranet (frequency) P 5.3 (0.7) .85** .72 .89* .62 .51T 4.4 (1.1) -.71 -.64

Databases (importance) P 4.8 (1.6) .80** .78* .66* .86* .65* .69* .82**

T 5.1 (1.9) .79* .63 .60Databases (frequency) P 4.0 (2.1) .62 .66 .95**

T 4.0 (1.3) .94** .68* .56 .63Access to IP (importance) P 2.8 (2.0) .54 .57 .67*

T 3.0 (2.2) .62 .80* .74*

Access to IP (frequency) P 1.5 (0.9) .73T 1.1 (0.4) .64

International research programs (importance) P 3.8 (1.6) .64*

T 4.9 (1.7) .75* .67International research programs (frequency) P 3.3 (1.3) .87** .94** .63* .61 .82* .72* .64* .68* .79*

T 3.6 (0.9) -.80Summit (importance) P 3.8 (1.8) .65* .66* .54 .79** .71*

T 5.7* (1.7) .60 .78*

Infrastructure (importance) P 3.9 (1.8) .83** .77* .75* .84* .66* .87** .92** .96**

T 4.1 (2.2) .59 .74* .74* .64

8P.J.P.

Garbade

et al.

/ N

JAS

- W

ageningen Journal

of Life

Sciences 67 (2013) 1– 10

Table 2 (continued )

Indicator Mean (Stdv) Knowledge valorization support

1 2 3 4 5 6 7 8 9 10 11

Infrastructure (frequency) P 3.0 (1.8) .75* .67 .59T 2.2 (1.7) .62

Technology monitoring (importance) P 4.5 (1.9) .75* .69* .72*

T 2.7 (1.5)Technology monitoring (frequency) P 2.8 (0.9) .66 .55 .51

T 1.9 (0.9)Technology advice (importance) P 3.4 (1.8) .79** .91** .80** .75* .79 .83** .70* .73* .56

T 4.0 (1.6)Technology advice (frequency) P 2.3 (1.0)

T 2.7 (1.4) .72* .90**

IP filing (importance) P 1 (0)T 2.3 (2.2) .59 .67

IP filing (frequency) P 1.1 (0.4) .51 .70 .73 .63 .60 .51T 1 (0)

Bio-informatics (importance) P 4.5 (2.0) .74* .56 .76 .69* .74* .81** .89**

T 4.6 (2.0) .63Bio-informatics (frequency) P 3.0 (1.8) .52 .79 .68*

T 3.3 (1.5)Troubleshooting (importance) P 2.3 (1.3) .88* .88** .69* .59

T 1.7 (1.3) .59 .67Troubleshooting (frequency) P 2.1**(1.0)

T 1.0 (0)Sharing R&D costs (importance) P 6.1**(0.8)

T 3.4 (2.0) .67 .56 .67 .57 .56Recruiting new researchers P 3.9 (2.1) .82** .97** .69* .76* .87* .72* .68* .73* .70*

T 4.0 (2.2) .67 .71*

Company interaction (importance) P 4.0 (2.4) .81* .92* .89 .72*

T 3.6 (1.7) .60 .62Company interaction (frequency) P 4.4* (0.8)

T 2.7 (1.5) .60 .74*

Bilateral research (importance) P 2.6 (1.8) .92** .67* .61 .60T 3.1 (1.2) .72 .66 .85**

Bilateral research (frequency) P 1.9 (1.2) .69 ,85* ,86** ,76* .58T 1.7 (1.0)

Extra research with knowledge centers P 6.3% (11%) ,73* .84* .81* .90* .69* .63* .81**

T 0.1% (0.2%)Extra research with commercial partners P 2.4% (3%) .80 .67 .63

T No extra res.

Shaded = questions with significant mean differences (1-tailed) between potato and tomato companies are shaded grey.Blank = no correlation at least at the 0.1 level (1-tailed); No star = Correlation is significant at the 0.1 level (1-tailed)

** Correlation is significant at the 0.05 level (1-tailed); ***Correlation is significant at the 0.01 level (1-tailed)a A number of breeding specific knowledge valorization performance questions (number 2,4,5,6) were not applicable to the potato processing companies, which reduces the potato companies sample to 6 companies for these

questions

en Jou

tcciocacittirsc‘ftsrp‘Ftdifrwpbr‘iwfiwwsbacarkfotwtoc‘a

ttcmCcCdsv

P.J.P. Garbade et al. / NJAS - Wagening

he tomato companies this increase also correlates with the indi-ator ‘breeding strategy improvement’. This last result seems inontrast to the higher importance potato companies give to thisndicator. A correlation between the indicator ‘new products devel-ped’ and ‘tested markers’ is found for both the tomato and potatoompanies. The valorization level correlates positively for tomatond potato companies with the indicator ‘basic research’, ‘suc-essful research completion’, ‘breeding strategy’ and ‘strengthenedmage’. Correlations with the indicator ‘new products launchedo the market’ are found for potato companies only. This reflectshe nature of the more applied projects potato companies engagen within CBSG. When relating the parameter ‘knowledge valo-ization performance’ to the parameter ‘knowledge valorizationupport’ by CBSG correlations are found for both tomato and potatoompanies between the indicators ‘basic research’ and ‘database’,infrastructure importance’ and ‘database frequency’, as well asor the ‘bio-informatics knowledge importance’. When looking athe indicator ‘plant breeding’, the potato and tomato companieshow significant correlations with the indicator ‘contact with CBSGesearchers (importance)’. The indicator ‘successful research com-letion’ correlates for both tomato and potato companies with

importance of database access’ and ‘access to the CBSG summit’.or tomato companies, correlations with the indicator ‘impor-ance of CBSG researchers contact’ and with the ‘frequency ofatabase and intellectual property access’ were also found. The

ndicator ‘time reduction in the breeding process’ is importantor the potato companies with a high number of positive cor-elations found with knowledge valorization support items. Thisas assessed to be of no importance by the tomato companieserhaps because these already have large in-house molecularreeding capacities. The indicator ‘tested markers’ is found cor-elated for both tomato and potato companies to the indicatorimportance of CBSG infrastructure’ and ‘help with troubleshoot-ng’. For the potato companies, it is also specifically correlated

ith the indicators ‘technology monitoring’ and road mappingrequency’, and with ‘importance of CBSG intranet access’. Thendicator ‘new products developed’ due to the CBSG participation

as found to correlate, for both potato and tomato companies,ith the indicators ‘access to intellectual property’, ‘CBSG infra-

tructure’ and ‘troubleshooting importance’. However it has toe borne in mind that tomato companies engage in CBSG moret the fundamental research level, which also explains negativeorrelations found between indicator ‘new products developed’nd ‘new products launched’ and frequency of ‘contact to CBSGesearchers’. Regarding the indicator ‘improvement of the level ofnowledge and skills of companies’ personnel’, correlations wereound for the potato companies with the indicators ‘importancef contact with CBSG researchers’, ‘access to intranet’, ‘interna-ional research programs frequency’, ‘importance of interactionith other companies’ and ‘the recruitment of new researchers’. For

he tomato companies a correlation with the indicator ‘frequencyf the database use’ should be mentioned. The valorization level isorrelated for both tomato and potato companies to the indicatorimportance of databases’ and ‘recruitment of new researchers andssistants’.

The differences found between the answers of the potato andomato companies in the survey give an interesting insight intoheir different expectation patterns in CBSG. The short PGLC tomatoompanies give mainly a high priority to obtain up-to-date infor-ation, as they judge the importance of the access to the annual

BSG summit and to the databases very high. The long PGLC potatoompanies give extra credits to the communication tool offered by

BSG in the form of technology monitoring and road mapping anderive extra value from accessing the CBSG infrastructure as can beeen from 8 positive correlations with the parameter ‘knowledgealorization performance’.

rnal of Life Sciences 67 (2013) 1– 10 9

6. Discussion and Conclusions

From the results of our in-depth investigation of the public pri-vate research partnership (PPRP) CBSG, it can be concluded thatsuch a partnership indeed increases the knowledge valorizationlevel. In general companies showed a high appreciation of theaccess to knowledge, such as the contact to CBSG researchers,the database, intranet and the CBSG summit meeting, i.e. theexchange of information. They also appreciate particularly some ofthe CBSG valorization support activities, such as technology mon-itoring and road mapping, the provided bio-informatics servicesand network growth. Besides this direct appreciation by the com-panies, all elements of knowledge valorization support were foundrelated to a higher knowledge valorization performance. Therefore,it is concluded that the research explorative strength of the publicinstitutes combined with the exploitative strength of the privateorganizations results in a tangible higher level of knowledge valo-rization performance. At the same time, thorough analysis of ourfindings gives ample indications that the type of valorization isalso affected by the different needs of the companies, in relationto the different lengths of their PGLC. Since companies with differ-ent PGLC lengths benefited from CBSG participation, it can also beconcluded that the length in PGLC is not a restriction to participatein and benefit from a PPRP. However, according to the length ofthe PGLC, the knowledge transfer as part of the knowledge valo-rization process takes place in different ways. Long PGLCs requireextra communication tools that focus on the long term R&D process.Companies with short term PGLCs are challenged by the race fornew products and not to miss out on opportunities. Consequently,the PPRP is valued here more for networking possibilities and as aprovider of the latest technological developments. Potato compa-nies clearly profit from CBSG as a PPRP in terms of their knowledgevalorization performance. Tomato companies, with their higher in-house molecular breeding capabilities, value CBSG more for thecontact with CBSG researchers and access to the annual CBSG sum-mit. For both potato and tomato companies, the contact with CBSGresearchers was found to be an important factor of the knowledgevalorization process, and was related to a higher knowledge valori-zation performance. Potato companies further derive their benefitsin the knowledge valorization process from gaining access to CBSGinfrastructure, intranet and databases and indicated a successfulknowledge transfer. Tomato companies seem to benefit also fromextending their need for research in collaboration with other com-panies within CBSG. Although they stated to be rather indifferenttowards the importance of the possibility of sharing R&D cost withother companies (conducting research collectively with CBSG part-ners), based on the significant correlations found, it seems thatthe cooperation aspect plays a major role for tomato companies aswell. Tomato companies especially benefit through the enhancedcompany cooperation in the CBSG precompetitive research. Thelong-term focus of precompetitive research appears to complementtheir daily business, the development of new tomato varieties. Atime reduction in the breeding process applies especially to thelong PGLC potato companies, which makes the PPRP, for them, ahighly effective means of knowledge valorization.

In conclusion, in contrast to the PPRP efficiency doubt raised bya number of authors, as mentioned in the introduction [e.g. 10], itcan be concluded that both potato and tomato companies benefitfrom their participation in CBSG. Furthermore, the general concernsraised by Adams [14] about PPRPs - that it would take up to 20 yearsto transform fundamental research in a way that industry can profitfrom it, can be refined by this study. The PGLC length reduction

achieved due to the participation in a PPRP like CBSG will shortenthe time that society will have to wait for new products. This canbe seen as the pay back to the tax payers’ money invested, and isan additional benefit to the outstanding scientific results obtained

1 en Jou

ifr

A

cdgSof2inwwO

R

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

in Patent Rights and Plant Breeder’s Rights, Wageningen, The Netherlands,

0 P.J.P. Garbade et al. / NJAS - Wagening

n CBSG as well as to its support in educating young science pro-essionals well suited to work either in academia or a commercialesearch environment.

cknowledgements

The authors thank the respondents from the CBSG participatingompanies for their willingness to fill in the questionnaires and toiscuss the findings in the interviews. In addition, the authors arerateful to Prof. Hans Dons for his precious advice and support, toebastian Sanchez Gerritsen for research assistance and to the teamf BICORE for collecting the data. The study has received fundingrom the European Union Seventh Framework Program (FP7/2007-013) under grant agreement n◦ 245301 NetGrow- “Enhancing the

nnovativeness of food SMEs through the management of strategicetwork behavior and network learning performance”. This projectas (co)financed by the Centre for BioSystems Genomics (CBSG),hich is part of the Netherlands Genomics Initiative/Netherlandsrganization for Scientific Research.

eferences

[1] G. Johnson, K. Scholes, R. Whittington, Exploring corporate strategy: text &cases, Prentice Hall, 2008.

[2] R. Coombs, M. Harvey, B.S. Tether, Analysing distributed processes ofprovision and innovation, Industrial and Corporate Change 12 (2003)1125–1155.

[3] W.W. Powell, K.W., Koput, L. Smith-Doerr, Interorganizational collaborationand the locus of innovation: Networks of learning in biotechnology, Adminis-trative science quarterly, (1996) 116-145.

[4] H. Chesbrough, W. Vanhaverbeke, J. West, Open Innovation Researching a NewParadigm: Researching a New Paradigm, OUP Oxford, 2008.

[5] D.B. Audretsch, A.N. Link, J.T. Scott, Public/private technology partnerships:evaluating SBIR-supported research, Research Policy 31 (2002) 145–158.

[6] C. Skelcher, Public-private partnerships and hybridity, The Oxford handbook ofpublic management, 2005, pp. 347–370.

[7] M. Perkmann, K. Walsh, How firms source knowledge from universities: part-nering versus contracting, in: J. Bessant, T. Venables (Eds.), Creating WealthFrom Knowledge, Meeting the Innovation Challenge, Edward Elgar Publishing,2008, pp. 273–296.

[8] M. Perkmann, K. Walsh, University–industry relationships and open innova-tion: Towards a research agenda, International Journal of Management Reviews9 (2007) 259–280.

[9] J. Goorden, R. v. Lieshout, E. Wubben, S. Omta, Towards a Classification of Instru-ments for Valorisation of Academic & Industrial Knowledge: an exploratoryanalysis of eight European incubators in the life sciences, High TechnologySmall Firms Conference, Enschede, The Netherlands, 2008.

10] E. Geisler, Creating value with science and technology, Greenwood PublishingGroup, 2001.

11] D. Schartinger, C. Rammer, M.M. Fischer, J. Fröhlich, Knowledge interactionsbetween universities and industry in Austria: sectoral patterns and determi-nants, Research Policy 31 (2002) 303–328.

12] P. D’Este, P. Patel, University–industry linkages in the UK: What are the factorsunderlying the variety of interactions with industry? Research Policy 36 (2007)1295–1313.

13] F. Meyer-Krahmer, U. Schmoch, Science-based technologies:university–industry interactions in four fields, Research Policy 27 (1998)835–851.

14] J.D. Adams, Fundamental stocks of knowledge and productivity growth, Journal

of Political Economy (1990) 673–702.

15] I. Feller, A historical perspective on government-university partnerships toenhance entrepreneurship and economic development, in: S. Shane (Ed.), Eco-nomic development through entrepreneurship: Government, university andbusiness linkages., Edward Elgar, Cheltenham, 2005, pp. 6–28.

[

rnal of Life Sciences 67 (2013) 1– 10

16] R. Fontana, A. Geuna, Factors Affecting University–Industry R&D Collaboration:The importance of screening and signalling, Bureau d’Economie Théorique etAppliquée, UDS, Strasbourg, 2005.

17] M.D. Santoro, A.K. Chakrabarti, Firm size and technology centrality inindustry–university interactions, Research Policy 31 (2002) 1163–1180.

18] R. Widdus, Public-private partnerships for health: their main targets, theirdiversity, and their future directions, Bulletin World Health Organization 79(2001) 713–720.

19] F. Fortuin, Strategic Alignment of Innovation to Business: Balancing Explorationand Exploitation in Short and Long Life Cycle Industries, Wageningen AcademicPub, 2007.

20] D. Lavie, L. Rosenkopf, Balancing Exploration and Exploitation in Alliance For-mation, Academy of Management Journal 49 (2006) 797–818.

21] D. Partha, P.A. David, Toward a new economics of science, Research Policy 23(1994) 487–521.

22] T.R. Behrens, D.O. Gray, Unintended consequences of cooperative research:Impact of industry sponsorship on climate for academic freedom and othergraduate student outcome, Research Policy 30 (2001) 179–199.

23] J.G. March, Exploration and exploitation in organizational learning, Organiza-tion Science 2 (1991) 71–87.

24] M. Perkmann, K. Walsh, The two faces of collaboration: impacts of university-industry relations on public research, Industrial and Corporate Change 18(2009) 1033–1065.

25] D.B. Audretsch, M. Keilbach, The Theory of Knowledge Spillover Entrepreneur-ship, Journal of Management Studies 44 (2007) 1242–1254.

26] E. Mansfield, Academic research underlying industrial innovations: sources,characteristics, and financing, The review of Economics and Statistics (1995)55–65.

27] W. Faulkner, J. Senker, L. Velho, Knowledge frontiers: public sector researchand industrial innovation in biotechnology, engineering ceramics, and parallelcomputing, Clarendon Press Oxford, 1995.

28] D. Hicks, Published papers, tacit competencies and corporate management ofthe public/private character of knowledge, Industrial and Corporate Change 4(1995) 401–424.

29] I.M. Cockburn, R.M. Henderson, Absorptive capacity, coauthoring behavior, andthe organization of research in drug discovery, The Journal of Industrial Eco-nomics 46 (1998) 157–182.

30] F. Murray, S. O’Mahony, Re-conceptualizing the institutional foundations ofcumulative innovation, Organization Science 18 (2007) 1006–1021.

31] B.L. Bayus, Are product life cycles really getting shorter? Journal of productinnovation management 11 (1994) 300–308.

32] W.E. Cox, Product life cycles as marketing models, The Journal of Business 40(1967) 375–384.

33] T. Levitt, Exploit the product life cycle, Graduate School of Business Adminis-tration, Harvard University, 1965.

34] R. Polli, V. Cook, Validity of the product life cycle, The Journal of Business 42(1969) 385–400.

35] G.A. Moore, Crossing the Chasm: Marketing and Selling High-Tech Products toMainstream Customers: HarperBusiness, 1999.

36] M.A. Maidique, B.J. Zirger, The new product learning cycle, Research Policy 14(1985) 299–313.

37] J. Williams, Renewable advantage: Crafting strategy through economic time,Free Press, 1999.

38] C.H. Fine, Clockspeed, Winning industry control in the age of temporary advan-tage, Basic Books, 1998.

39] S.L. Brown, K.M. Eisenhart, Competing on the edge: Strategy as structuredchaos, Boston, Mass, Harvard Business School Press, 1998.

40] OECD, Public–Private Partnerships for Research and Innovation: An Evalua-tion of the Dutch Experience., Organisation for Economic Co-operation andDevelopment, Paris, 2003.

41] J. Spiertz, M. Kropff, Adaptation of knowledge systems to changes in agricul-ture and society: The case of the Netherlands, NJAS-Wageningen journal of lifesciences 58 (2011) 1–10.

42] N. Louwaars, H. Dons, G. Van Overwalle, H. Raven, A. Arundel, D. Eaton, A. Nelis,Breeding Business The Future of Plant Breeding in the Light of Developments

2010.43] M. Perkmann, A. Neely, K. Walsh, How should firms evaluate success in

university–industry alliances? A performance measurement system, R&D Man-agement 41 (2011) 202–216.

http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0005http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0005http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0005http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0005http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0005http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0005http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0005http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0005http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0005http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0005http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0005http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0005http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0005http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0005http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0005http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0010http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0010http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0010http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0010http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0010http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0010http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0010http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0010http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0010http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0010http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0010http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0010http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0010http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0010http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0010http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0010http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0010http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0010http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0010http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0010http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0010http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0010http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0020http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0020http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0020http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0020http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0020http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0020http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0020http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0020http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0020http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0020http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0020http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0020http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0020http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0020http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0020http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0020http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0020http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0020http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0020http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0025http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0025http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0025http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0025http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0025http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0025http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0025http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0025http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0025http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0025http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0025http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0025http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0025http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0025http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0025http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0025http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0025http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0025http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0025http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0030http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0030http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0030http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0030http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0030http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0030http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0030http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0030http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0030http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0030http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0030http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0030http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0030http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0030http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0030http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0030http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0030http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0035http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0035http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0035http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0035http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0035http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0035http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0035http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0035http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0035http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0035http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0035http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0035http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0035http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0035http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0035http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0035http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0035http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0035http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0035http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0035http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0035http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0035http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0035http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0035http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0035http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0035http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0035http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0035http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0035http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0035http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0035http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0035http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0035http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0035http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0035http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0035http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0040http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0040http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0040http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0040http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0040http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0040http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0040http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0040http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0040http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0040http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0040http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0040http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0040http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0040http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0040http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0040http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0040http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0040http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0040http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0040http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0040http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0040http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0040http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0040http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0040http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0040http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0045http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0045http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0045http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0045http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0045http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0045http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0045http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0045http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0045http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0045http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0045http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0045http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0045http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0045http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0045http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0045http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0045http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0045http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0045http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0045http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0045http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0045http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0045http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0045http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0045http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0045http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0045http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0045http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0045http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0045http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0045http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0045http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0045http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0045http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0045http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0045http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0045http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0045http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0045http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0045http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0045http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0045http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0050http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0050http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0050http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0050http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0050http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0050http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0050http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0050http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0050http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0050http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0050http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0050http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0055http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0055http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0055http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0055http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0055http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0055http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0055http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0055http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0055http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0055http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0055http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0055http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0055http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0055http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0055http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0055http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0055http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0055http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0055http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0055http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0055http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0055http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0055http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0055http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0055http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0055http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0055http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0055http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0060http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0060http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0060http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0060http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0060http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0060http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0060http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0060http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0060http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0060http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0060http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0060http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0060http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0060http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0060http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0060http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0060http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0060http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0060http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0060http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0060http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0060http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0060http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0060http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0060http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0060http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0060http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0060http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0060http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0065http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0065http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0065http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0065http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0065http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0065http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0065http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0065http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0065http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0065http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0065http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0065http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0065http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0065http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0065http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0065http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0065http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0065http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0065http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0065http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0070http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0070http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0070http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0070http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0070http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0070http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0070http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0070http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0070http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0070http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0070http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0070http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0070http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0070http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0070http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0070http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0070http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0075http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0075http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0075http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0075http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0075http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0075http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0075http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0075http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0075http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0075http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0075http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0075http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0075http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0075http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0075http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0075http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0075http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0075http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0075http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0075http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0075http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0075http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0075http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0075http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0075http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0075http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0075http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0075http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0075http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0075http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0075http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0075http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0075http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0075http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0075http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0075http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0085http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0085http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0085http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0085http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0085http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0085http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0085http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0085http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0085http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0085http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0085http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0085http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0085http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0085http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0085http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0085http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0085http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0085http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0085http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0085http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0085http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0090http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0090http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0090http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0090http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0090http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0090http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0090http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0090http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0090http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0090http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0090http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0090http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0090http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0090http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0090http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0090http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0090http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0090http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0090http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0090http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0090http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0090http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0090http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0090http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0095http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0095http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0095http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0095http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0095http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0095http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0095http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0095http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0095http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0095http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0095http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0095http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0095http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0095http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0095http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0095http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0095http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0095http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0095http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0095http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0095http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0095http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0095http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0100http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0100http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0100http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0100http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0100http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0100http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0100http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0100http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0100http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0100http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0100http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0100http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0100http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0100http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0100http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0100http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0100http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0100http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0100http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0100http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0100http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0105http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0105http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0105http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0105http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0105http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0105http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0105http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0105http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0105http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0105http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0105http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0105http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0105http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0105http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0105http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0105http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0105http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0110http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0110http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0110http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0110http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0110http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0110http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0110http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0110http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0110http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0110http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0110http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0110http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0110http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0110http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0110http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0110http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0110http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0110http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0110http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0110http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0110http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0110http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0110http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0110http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0110http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0110http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0110http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0110http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0110http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0110http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0115http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0115http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0115http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0115http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0115http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0115http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0115http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0115http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0115http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0115http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0115http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0115http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0115http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0115http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0115http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0115http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0120http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0120http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0120http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0120http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0120http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0120http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0120http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0120http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0120http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0120http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0120http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0120http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0120http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0120http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0120http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0120http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0120http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0120http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0120http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0120http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0120http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0120http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0120http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0120http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0120http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0120http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0125http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0125http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0125http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0125http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0125http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0125http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0125http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0125http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0125http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0125http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0125http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0125http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0125http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0125http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0125http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0125http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0125http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0125http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0125http://refhub.elsevier.com/S1573-5214(13)00021-3/sbref0